A data management plan (DMP) is a digital document in which you describe what data you will collect during your research project, how you are going to store and manage the data during the project, and what will happen to the data after the project is finished.
A data management plan makes research data management concrete and practicable: what are you going to do and how?
It saves you time, work and money: by thinking out the management of research data at an early stage, you reduce the chances of having to face nasty surprises later in your research.
A growing number of universities, research institutes and funders make a (succinct) DMP mandatory.
In a data management plan you provide answers to such questions as:
A DMP is a living document. It may well be that you are not able to answer some of these questions at first, or only roughly, for example because you wish to obtain someone's advice. Or it may turn out that it is better to handle things differently than originally planned. You can always adapt or add to your DMP.
There are various templates and checklists available for drawing up a data management plan. The filled-in template is then your data management plan. Your research funder or research institute may have its own DMP template. If you receive funding from NWO or ZonMw, or if your funder or institute does not provide a template, you can use the general UvA or AUAS template. Many templates, among which the UvA and AUAS ones, are available in the online tool DMPonline.
If you apply for funding at one of the English Research Councils, you can use DMPonline to write a data management plan as well. Templates which satisfy the requirements of the most important American research funders are available via the DMPTool.
Good data management involves costs, both in hours and in money. The UK Data Service has developed a checklist which can help you work out these costs, for example for a grant request.